•  
  •  
 

Keywords

Drainage robotic, machine learning, IoT, YOLOv8, solar power, object detection

Document Type

Research Article

Abstract

Drain blockages caused by dried leaves and debris are a recurring issue in residential areas, particularly in tropical regions such as Malaysia leading to hygiene risks and increased manual maintenance effort. This paper presents the design and implementation of a solar-powered autonomous drain-cleaning robot for domestic applications. The system integrates a mobile robotic platform with a 4-DOF robotic arm, YOLOv8-based vision for real-time debris detection, and Internet of Things (IoT)-based remote monitoring. The robot is built on a Raspberry Pi 5 and incorporates VL53L0X time-of-flight (ToF) sensors, INA219 power sensors, NEO-6M GPS and a USB webcam. Detected debris triggers an automated scooping sequence executed by the robotic arm. Power is supplied by an 11.1 V, 5200 mAh LiPo battery, supplemented by a 12 V, 3.6 W solar panel, with telemetry transmitted via MQTT. Experimental results show that the YOLOv8 model achieved a mean Average Precision ([email protected]–0.95) of 88%, with stable operation at 2.0–3.5 frames per second (FPS) and approximately 1.12 h of continuous runtime per charge (57.72 Wh). Solar charging replenished 6.48% of battery capacity (3.74 Wh) within 2 h, reducing reliance on grid electricity. These results demonstrate a measurable, energy-efficient, and cost-effective solution that contributes to SDGs 6, 9, 11, and 13.

html

Publication Date

31-3-2026

First Page

1

Last Page

7

Download

Share

COinS